|Publication number||US7480760 B2|
|Application number||US 10/738,506|
|Publication date||Jan 20, 2009|
|Filing date||Dec 17, 2003|
|Priority date||Dec 17, 2003|
|Also published as||US20050138271, WO2005059966A2, WO2005059966A3|
|Publication number||10738506, 738506, US 7480760 B2, US 7480760B2, US-B2-7480760, US7480760 B2, US7480760B2|
|Inventors||Sacha Bernstein, Anthony L. Gelsomini|
|Original Assignee||Wegener Communications, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (18), Referenced by (9), Classifications (15), Legal Events (3)|
|External Links: USPTO, USPTO Assignment, Espacenet|
1. Field of the Invention
This invention relates generally to flash memory devices and, more particularly, to a method of rotating usage of memory partitions to extend the life of the flash memory device.
2. Related Art
Flash memory is a single transistor cell which is programmable through hot electron injection or source injection and erasable through Fowler-Nordheim tunneling. The programming and erasing of such a memory cell requires current to pass through the dielectric surrounding a floating gate electrode. Because of this, such types of memory have a finite number of erase-write cycles. Eventually, the dielectric will fail. Manufacturers of flash cell devices specify the limit for the number of erase-write cycles as between 10,000 and 100,000. Some flash technologies have over-erase problems, therefore they need to be programmed before erasing to improve uniformity. They also have to be verified for successful erase. Accordingly, unlike rotating magnetic media, a flash memory device does not have an indefinite lifetime.
Flash memory devices have many advantages for a large number of applications. These advantages include their non-volatility, speed, ease of erasure and reprogramming, small physical size and related factors. There are no mechanical moving parts and as a result such systems are not subject to failures of the type most often encountered with hard disk storage systems.
Flash memory devices have many characteristics that are different from other memory devices. One major difference is that a block containing existing data in flash memory devices cannot be overwritten with new data. Existing data must be completely erased (also referred to as “cleaned”) from a block before data can be written into memory locations again. For the reasons noted above, blocks can be erased only a limited number of times before the flash memory device becomes unusable.
Additionally, most block erasures stall other operations from occurring such as read and write operations to the flash memory device. Most flash memory systems attempt, therefore, to minimize erasures to specific times, such as at initialization or powering-off of a computer system, or at periodically scheduled times.
It is in view of the above problems that the present invention was developed. The invention is a data structure, system and method to minimize the number of erase cycles performed on a flash memory device to extend its useful life. A flash memory device has several areas where data is stored. Each area is referred to as a block. Memory usage is rotated between blocks to evenly distribute erase cycles.
In one aspect of the invention, there are two blocks. Data is stored in the first block until it becomes full. When the first block is full, the memory storage area is rotated to the second block. Data is stored in the second until it becomes full. The first block is erased and the process repeats.
In another aspect of the invention, data is stored in a memory area consecutively, much like one would write in a journal. Each memory area is similar to a page and data is stored on the page much like a journal entry. A flash driver directs data to be written on the page until that page is full. When the current page is full, the flash driver directs data to a subsequent page.
In yet another aspect of the invention, there is a dual-function node header. The node header stores information which may be used to identify a valid record. The node header also stores information regarding status of a particular block. For example, the node header can indicate whether the particular block is full, partially written to, or erased. The node header eliminates the need for a look-up table as found in the prior art.
Further features and advantages of the present invention, as well as the structure and operation of various embodiments of the present invention, are described in detail below with reference to the accompanying drawings.
The accompanying drawings, which are incorporated in and form a part of the specification, illustrate the embodiments of the present invention and together with the description, serve to explain the principles of the invention. In the drawings:
Referring to the accompanying drawings in which like reference numbers indicate like elements,
Universal Flash Medium Operating Characteristics
Each block, 0, 1, 2, etc., is further subdivided into K sectors 102; standard commercial NAND flash media commonly contain 8, 16, or 32 sectors per block. The amount of blocks and sectors can vary, however, depending on the manufacturer. Some manufacturers refer to “sectors” as “pages.” Both terms as used herein are equivalent and interchangeable.
Each sector 102 is further divided into two distinct sections, a data area 103 used to store information and a spare area 104 which is used to store extra information such as error correction code (ECC). The data area 103 size is commonly implemented as 512 bytes, but again could be more or less depending on the manufacturer. At 512 bytes, the flash memory medium allows most file systems to treat the medium as a nonvolatile memory device, such as a fixed disk (hard drive). As used herein RAM refers generally to the random access memory family of memory devices such as DRAM, SRAM, VRAM, VDO, and so forth. Commonly, the size of the spare area 104 is implemented as 16 bytes of extra storage in NAND flash media devices. Again, other sizes, greater or smaller can be selected. In most instances, the spare area 104 is used for error correcting codes, and status information.
A NOR memory medium 200 is different than a NAND memory medium in that blocks are not subdivided into physical sectors, but are subdivided into cells 210. Each cell 210 is one byte in length. Similar to RAM, each byte stored within a block of NOR memory medium is individually addressable.
Aside from the overall layout and operational comparisons, some universal electrical characteristics (also referred to herein as “memory requirements” or “rules”) of flash devices can be summarized as follows:
Computer device 300 generally includes a processor 302, memory 304, and a flash memory media 100/200. The computer device 300 can include more than one of any of the aforementioned elements. Other elements such as power supplies, keyboards, touch pads, I/O interfaces, displays, LEDs, audio generators, vibrating devices, and so forth are not shown, but could easily be a part of the exemplary computer device 300.
Memory 304 generally includes both volatile memory (e.g., RAM) and non-volatile memory (e.g., ROM, PCMCIA cards, etc.). In most implementations described below, memory 304 is used as part of computer device's 300 cache, permitting application data to be accessed quickly without having to permanently store data on a non-volatile memory such as flash medium 100/200.
An operating system 309 is resident in the memory 304 and executes on the processor 302. An example operating system implementation includes the Windows® CE operating system from Microsoft Corporation, but other operation systems can be selected from one of many operating systems, such as DOS, UNIX, LINUX, etc. For purposes of illustration, programs and other executable program components such as the operating system are illustrated herein as discrete blocks, although it is recognized that such programs and components reside at various times in different storage components of the computer, and are executed by the processor(s) of the computer device 300.
One or more application programs 307 are loaded into memory 304 and run on the operating system 309. Examples of applications include, but are not limited to, email programs, word processing programs, spreadsheets programs, Internet browser programs, as so forth.
Also loaded into memory 304 is a file system 305 that also runs on the operating system 309. The file system 305 is generally responsible for managing the storage and retrieval of data to memory devices, such as magnetic hard drives, and also, in this exemplary implementation flash memory media 100/200. Most file systems 305 access and store information at a logical level in accordance with the conventions of the operating system the file system 305 is running. It is possible for the file system 305 to be part of the operating system 309 or embedded as code as a separate logical module.
Flash driver 306 is implemented to function as a direct interface between the file system 305 and flash medium 100/200. Flash driver 306 enables computer device 300 through the file system 305 to control flash medium 100/200 and ultimately send/retrieve data. As shall be described in more detail, however, flash driver 306 is responsible for more than read/write operations. Flash driver 306 is implemented to maintain data integrity, minimize data loss during a power interruption to computer device 300 and permit OEMs of computer devices 300 to support their respective flash memory devices regardless of the manufacturer. The flash driver 306 is file system agnostic. That means that the flash driver 306 supports many different types of files systems, such as File Allocation Data structure File System (FAT16), (FAT32), and other file systems. Additionally, flash driver 306 is flash memory medium agnostic, which likewise means driver 306 supports flash memory devices regardless of the manufacturer of the flash memory device. That is, the flash driver 306 has the ability to read/write/erase data on a flash medium and can support most, if not all, flash devices.
In the exemplary implementation, flash driver 306 resides as a module within operating system 309, that when executed serves as a logical interface module between the file system 305 and flash medium 100/200. The flash driver 306 is illustrated as a separate box 306 for purposes of demonstrating that the flash driver when implemented serves as an interface. Nevertheless, flash driver 306 can reside in other applications, part of the file system 305 or independently as separate code on a computer-readable medium that executes in conjunction with a hardware/firmware device.
In one implementation, flash driver 306 includes: a flash abstraction logic 308 and a programmable flash medium logic 310. Flash abstraction logic 308 and programmable medium logic 310 are coded instructions that support various features performed by the flash driver 306. Although the exemplary implementation is shown to include these two elements, various features from each of the flash abstraction logic 308 and flash medium logic 310 may be selected to carry out some of the more specific implementations described below. So while the described implementation shows two distinct layers of logic 308/310, many of the techniques described below can be implemented without necessarily requiring all or a portion of the features from either layer of logic. Furthermore, the techniques may be implemented without having the exact division of responsibilities as described below, all without departing from the scope of the present invention.
The programmable medium logic 310 includes a programmable entry point module (not shown), I/O module (not shown), and an ECC module (not shown). The programmable entry point module defines a set of programming interfaces to communicate between flash abstraction logic 308 and flash medium 100/200. In other words, the programmable entry points permit manufacturers of computer devices 300 to program the flash media logic 310 to interface with the actual flash memory medium 100/200 used in the computer device 300. The I/O module contains specific code necessary for read/write/erase commands that are sent to the Flash memory medium 100/200. The user can program the ECC module to function in accordance with any particular ECC algorithm selected by the user.
In one implementation, the flash abstraction logic 308 manages those operating characteristics that are universally common to flash memory media. These universal memory requirements include maintaining data integrity and handling recovery of data after a power failure.
The flag 410 stores information on the flash memory block in which data is stored. For example, if the flag 410 is “FF” in hexadecimal, the block is completely erased and ready to be written. However, if the flag 410 is “7F” in hexadecimal, then the block is at least partially written. Finally, if the flag 410 is “00” in hexadecimal, then all segments or cells have been written, and the block may need to be erased.
In the present invention, the pointer 412 identifies a current valid record. Valid means the most recent record. For example, if the flash medium 100/200 is used to store configuration records, then the valid record is the configuration record that was written last. As an example, the pointer 412 provides the bit length to the next available record. The pointer 412 indicates whether a particular segment or cell has data and whether the data is valid or invalid.
In one embodiment of the invention, data records are written to the flash medium 100/200 in convention much like one records entries into a journal or diary. In other words, the flash medium 100/200 has discrete sections that are consecutively arranged. One of the discrete sections is identified as the current memory page. Data records are written to the current page until that page is full. When the page is full, the next consecutive page is identified as the current memory page. The node header 450 contains information regarding the data record and may contain information regarding the block. Continuing with the analogy, the node header 450 acts like a journal entry date and may include a journal page number.
Suppose a third record is to be written in the spaced indicated as 460. First, a third header 462 will be written by the processor 302. The third header 462 will be written from “FFFF” to “0x7FFF.” This changes the first bit of the flag from a binary “1” to zero. A third data 464 will be written consecutively after the third header. Next, location, or “journal entry,” must be given to the third header 462. As such, the second header 442 is rewritten to store the second data length, “132” for example. In this way, the second header 442 indicates the position of the third header 462.
If so, the processor 302 proceeds on to the next block in step 505. In an optional step, the processor may write the first two bytes of the instant block to all zeroes, thereby acting as a flag to indicate the particular block is full. The processor then checks to see if there is sufficient room in the current block to write the data string in step 516. If there is not sufficient room to write the current data string, then the processor 302 moves on to the next block.
If so, then the processor writes the data string to the identified next available segment. The processor 302 generates a header having a flag and a pointer as described above in conjunction with
If the block has been flagged, or if the block is full, or if there is insufficient room to write the data string, then the processor moves on to the next block in step 505. For example, if the flash medium 100/200 has two blocks, the processor would stop looking at block 1 and would look in block 2. As shown in
The processor 302 searches for the pointer indicating a valid current record in step 716. The processor 302 searches for the valid record by reading each header of the current block. The processor 302 starts in the first header of the first block. For example, if the first header reads “0x7FFF” in hexadecimal, then the data following the header is the valid record. However, if the header indicates some other value, then that value leads to the next data record. The processor 302 then sequentially reads each header to determine which header is adjacent the valid data record.
The processor 302 decides whether the pointer indicating a valid record is found in step 718. If so, the processor 302 reads the data string in step 720. If a pointer is not found in the current block, the processor 302 increments the block number by one in step 722 thereby directing the processor 302 to the next block. The processor 302 then begins the search again and repeats this process until the pointer indicating a valid record is found.
As an example, and referring once again to
In view of the foregoing, it will be seen that the several advantages of the invention are achieved and attained.
The embodiments were chosen and described in order to best explain the principles of the invention and its practical application to thereby enable others skilled in the art to best utilize the invention in various embodiments and with various modifications as are suited to the particular use contemplated.
As various modifications could be made in the constructions and methods herein described and illustrated without departing from the scope of the invention, it is intended that all matter contained in the foregoing description or shown in the accompanying drawings shall be interpreted as illustrative rather than limiting. For example, the node header is illustrated as having 16 bits but may have 24 bits. Thus, the breadth and scope of the present invention should not be limited by any of the above-described exemplary embodiments, but should be defined only in accordance with the following claims appended hereto and their equivalents.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5341339||Nov 1, 1993||Aug 23, 1994||Intel Corporation||Method for wear leveling in a flash EEPROM memory|
|US5479638 *||Mar 26, 1993||Dec 26, 1995||Cirrus Logic, Inc.||Flash memory mass storage architecture incorporation wear leveling technique|
|US5485595||Oct 4, 1993||Jan 16, 1996||Cirrus Logic, Inc.||Flash memory mass storage architecture incorporating wear leveling technique without using cam cells|
|US5530827||Nov 24, 1992||Jun 25, 1996||Fujitsu Limited||Data management system for programming-limited type semiconductor memory and IC memory card employing data save/erase process with flag assignment|
|US5644539||Nov 25, 1992||Jul 1, 1997||Hitachi, Ltd.||Storage device employing a flash memory|
|US5835935||Sep 13, 1995||Nov 10, 1998||Lexar Media, Inc.||Method of and architecture for controlling system data with automatic wear leveling in a semiconductor non-volatile mass storage memory|
|US5983312||Aug 18, 1997||Nov 9, 1999||Fujitsu Limited||Simultaneously writing to and erasing two commonly numbered sectors|
|US6405279||Jun 10, 1999||Jun 11, 2002||Denso Corporation||Apparatus and method for controlling rewriting of data into nonvolatile memory|
|US6412080 *||Feb 23, 1999||Jun 25, 2002||Microsoft Corporation||Lightweight persistent storage system for flash memory devices|
|US6615318||Jan 22, 2002||Sep 2, 2003||International Business Machines Corporation||Cache management system with multiple cache lists employing roving removal and priority-based addition of cache entries|
|US6621746||Feb 27, 2002||Sep 16, 2003||Microsoft Corporation||Monitoring entropic conditions of a flash memory device as an indicator for invoking erasure operations|
|US20030041210||Aug 24, 2001||Feb 27, 2003||Micron Technology, Inc.||Erase block management|
|US20030135688||Apr 8, 2002||Jul 17, 2003||Shih-Chieh Tai||Flash memory data access method and configuration|
|US20030149843||Jan 22, 2002||Aug 7, 2003||Jarvis Thomas Charles||Cache management system with multiple cache lists employing roving removal and priority-based addition of cache entries|
|US20030204665||Apr 26, 2002||Oct 30, 2003||Jain Raj Kumar||High performance architecture with shared memory|
|GB2305272A||Title not available|
|WO1995010083A1||Sep 23, 1994||Apr 13, 1995||Cirrus Logic, Inc.||Flash memory with reduced erasing and overwriting|
|WO1997010604A1||Sep 12, 1996||Mar 20, 1997||Lexar Microsystems Inc.||Method of and architecture for controlling system data with automatic wear leveling in a semiconductor non-volatile mass storage memory|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7876616 *||Nov 12, 2007||Jan 25, 2011||Cadence Design Systems, Inc.||System and method for wear leveling utilizing a relative wear counter|
|US7882299 *||Feb 1, 2011||Sandisk Corporation||System and method for use of on-chip non-volatile memory write cache|
|US8621328||Mar 4, 2011||Dec 31, 2013||International Business Machines Corporation||Wear-focusing of non-volatile memories for improved endurance|
|US8635397 *||Mar 23, 2011||Jan 21, 2014||Spansion Llc||Data writing method and system|
|US9142301 *||Jan 7, 2014||Sep 22, 2015||Cypress Semiconductor Corporation||Data writing method and system|
|US20060136656 *||Dec 21, 2004||Jun 22, 2006||Conley Kevin M||System and method for use of on-chip non-volatile memory write cache|
|US20090122949 *||Nov 12, 2007||May 14, 2009||Denali Software, Inc.||System and method for wear leveling utilizing a relative wear counter|
|US20110258517 *||Oct 20, 2011||Fujitsu Semiconductor Limited||Data writing method and system|
|US20140122785 *||Jan 7, 2014||May 1, 2014||Spansion Llc||Data writing method and system|
|U.S. Classification||711/103, 711/E12.008, 365/185.33, 711/156, 365/185.09|
|International Classification||G11C16/34, G06F12/16, G06F12/02, H01L, G11C11/34, G06F12/00|
|Cooperative Classification||G06F2212/7211, G11C16/349, G06F12/0246|
|Dec 17, 2003||AS||Assignment|
Owner name: WEGENER COMMUNICATIONS, INC., GEORGIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:BERNSTEIN, SACHA;GELSOMINI, ANTHONY L.;REEL/FRAME:014809/0463
Effective date: 20031212
|Jul 9, 2012||FPAY||Fee payment|
Year of fee payment: 4
|Sep 2, 2016||REMI||Maintenance fee reminder mailed|